Thinking about the "The Big OOPs: Anatomy of a Thirty-five-year Mistake" talk by Casey Muratori. There was something that I disagreed with while watching this on stream that I feel is worth noting. At one point in the talk he says...

And we could have had [discriminated unions] from day one if someone had just believed that this was good.

...after talking about C.A.R Hoare's paper 1966 Record Handling. However, the language ML was first introduced in 1973 and it has pattern matching. ML went on to influence successful functional programming languages like Haskell.

I suppose Casey means that he wishes it was picked up by the mainstream, but I think the framing of it seeming like the idea was lost at that point is a little unfair.

Perhaps it's possible that ML didn't have pattern matching until relatively recently; I'm not actually sure if ML was introduced with pattern matching or if not, when it was added exactly.